Risk prediction scores for postoperative mortality after esophagectomy: validation of different models

Area of Science:

• Surgical Oncology
• Outcomes Research
• Predictive Analytics

Background:

• Esophagectomy carries significant operative mortality risks.
• Several prediction models exist to estimate these risks.
• Independent validation of these models is crucial for clinical utility.

Purpose of the Study:

• To independently validate the predictive performance of the Philadelphia, Rotterdam, Munich, and American Society of Anesthesiologists (ASA) scores.
• To assess the generalizability of these models across different patient populations undergoing esophagectomy for cancer.

Main Methods:

• Logistic regression models were employed for validation.
• Two distinct cohorts of esophagectomy cancer patients from Switzerland (n=170) and Australia (n=176) were utilized.
• Model performance was evaluated for 30-day mortality and in-hospital death prediction.

Main Results:

• The Philadelphia and Rotterdam scores demonstrated significant predictive value for 30-day mortality and in-hospital death in the pooled cohort.
• The Philadelphia score was the only model with significant 30-day mortality prediction in both individual cohorts.
• Predictive performance varied significantly between cohorts, with the Munich score showing no significant predictive value.
• ASA score showed significant predictive value for 30-day mortality in the Swiss cohort and in-hospital death in the pooled and Swiss cohorts.

Conclusions:

• Current prediction models, including Philadelphia, Rotterdam, Munich, and ASA, lack universal applicability for esophagectomy operative mortality.
• Development of improved, generalizable predictive scores or country-specific models is recommended.
• Clinical decision-making requires careful consideration of regional variations in patient outcomes and model performance.